Finally, in area 3 we propose to perform detailed studies to charaderize the effeds of genetic variation on risk for lung cancer, allowing for effects from tobacco smoke and a limited set of additional environmental exposures that have been collected as a part of ILCCO. We will also explore the effects of genetic variation on smoking phenotypes. Area 3 analyses will allow a careful assessment of the joint effects of smoking and genetic exposures for a range of populations having different smoking behaviors, genetic backgrounds and effects from potential confounding fadors such as diet that have been measured in many cases, but with imprecision. In addition, we will be evaluating existing procedures for characterizing the risk for lung cancer according to age, sex, smoking behavior and demographic and medical history factors. We will also extend these models to include the results from the proposed genetic studies that we are conduding. The final product will be the development of a clearer view of lung carcinogenesis as well as better tools for identifying individuals at high risk to develop lung cancer. Identifying such people is an important goal, since there remain no approved screening mortalities for lung cancer in asymptomatic individuals because of a low positive predidive value of such approaches in the general population, and concern about the potential risks and costs associated with screening.